10 H. F. Blanford— r/te Theory of the Winter [No. 1, 



and on the hills of the Northern Punjab that these rains are most copious, 

 and that the maximum falls latest ; the precipitation frequently taking 

 the form of snow at all but the less elevated stations. In the extreme 

 north-west, they attain their maximum in March and April ; for the 

 April rainfall (not shewn in the table) is about equal to that of March 

 at Peshawar and in the hills of Hazara, while that of May and June 

 is insignificant. But south of the Salt Range, and on the plains of the 

 Eastern Punjab, the rainfall of February and March exceeds that of 

 April. Still further to the south-east, throughout the greater part of 

 the Gangetic plain, the maximum occurs still earlier, viz., in January ; 

 and this holds good as far as Behar and the confines of Northern Ben- 

 gal. This anticipation of the maximum is not due to the January 

 rainfall of the Gangetic plain being heavier than that of the Punjab. 

 On the contrary, it is rather less : but the decrease in a south and S. E. 

 direction is muc?i less rapid in January than in the subsequent months. 

 In Northern Bengal, even the January maximum has vanished ; and 

 while the average rainfall of that month is only slightly less than in 

 Behar and the eastern part of the N. W. Provinces, that of February 

 is higher, and that of March shews a further considerable increase. So 

 far, the course of the variation seems to resemble that of the N. "W. 

 Punjab ; but the further steady increase of the fall in April, May, and 

 June shews that this resemblance is fallacious, and that we have here 

 to do with a phenomenon of a different order, viz., the storm precipita- 

 tion of the spring months, the characteristics of which are still more 

 pronounced in the more easterly province of Assam. As a well-marked 

 feature of the local meteorology, the cold weather rainfall does not 

 extend, in an easterly direction, beyond the province of Behar. 



Turning now to the regions somewhat further south, but still, for 

 the most part, without the tropic, we see that in Upper Sind the total 

 fall of the five months is very small, notwithstanding that it represents 

 nearly one-third of that of the year ; and also that, as in the Derajat, 

 it reaches its maximum in March. In Lower Sind, Cutch, and Gujarat, 

 the whole precipitation of the season is insignificant ; but a maximum 

 is still faintly indicated in February, and the same is more strongly 

 marked in Rajputana, where the fall is about double as great. In 

 Rajputana, it would appear that the December rainfall is almost as 

 great as that of March ; but this is mainly due to an exceptionally heavy 

 rainfall in December 1877,* combined with the fact that the Rajputana 

 reo-isters extend over a shorter period than those of most other parts 



* On this occasion nearly five inches of rain were registered at Banswara, 

 between 3 and 4 inches at Jhalrapatam and Ulwar, and over "2 inches at Kotah, 

 Deoli, and Bhurtpore. 



